CN114800500B - Flexible constant force control method and system for polishing robot - Google Patents
Flexible constant force control method and system for polishing robot Download PDFInfo
- Publication number
- CN114800500B CN114800500B CN202210422569.1A CN202210422569A CN114800500B CN 114800500 B CN114800500 B CN 114800500B CN 202210422569 A CN202210422569 A CN 202210422569A CN 114800500 B CN114800500 B CN 114800500B
- Authority
- CN
- China
- Prior art keywords
- information
- compensation
- attitude
- constant force
- grinding
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J11/00—Manipulators not otherwise provided for
- B25J11/005—Manipulators for mechanical processing tasks
- B25J11/0065—Polishing or grinding
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1628—Programme controls characterised by the control loop
- B25J9/1633—Programme controls characterised by the control loop compliant, force, torque control, e.g. combined with position control
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1679—Programme controls characterised by the tasks executed
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Manipulator (AREA)
Abstract
The invention provides a flexible constant force control method and a system for a polishing robot, wherein the method is applied to the field of artificial intelligence, and comprises the following steps: acquiring first grinding information of a first grinding robot; obtaining constant force compensation information according to the first grinding information; carrying out constant grinding force compensation adjustment on the first grinding robot, and carrying out posture detection to obtain first posture information; constructing a posture balance compensation state space; acquiring various attitude compensation behavior information; constructing attitude compensation optimization fitness, and optimizing in various attitude compensation behavior information; and performing attitude balance compensation by adopting the optimal attitude compensation behavior information to polish, and taking the compensated attitude information as initial attitude information for performing attitude balance compensation next time. The control precision and the production workpiece quality of the polishing robot are improved, the safety problem occurrence probability in the polishing process is reduced, and the technical effect of accurate control is achieved.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a flexible constant force control method and system for a polishing robot.
Background
The traditional robot only has the displacement concept in the grinding and polishing operation, but for the contact operation such as polishing, grinding, assembling and the like, the requirements of machining precision and machining quality cannot be met by only adopting position control. In order to ensure the processing quality of a workpiece during grinding and polishing and prevent the robot or the workpiece from being damaged when the end effector is in contact with the workpiece, the robot needs to be effectively controlled by force.
The technical problems that in the prior art, the precision of a position control method adopted in the robot polishing operation process is low, the workpiece polishing quality is poor, and the safety of a polishing robot and the workpiece has certain hidden danger exist.
Disclosure of Invention
The application provides a flexible constant force control method and system for a polishing robot, and solves the technical problems that in the prior art, the precision of a position control method adopted in the polishing operation process of the robot is low, the polishing quality of a workpiece is poor, and certain hidden danger exists in the safety of the polishing robot and the workpiece. The method has the advantages that the constant force compensation analysis model is constructed by collecting the polishing operation parameter information of the robot, the constant force compensation information is generated in a targeted manner, and the attitude optimization is carried out according to the compensated attitude, so that the control precision and the quality of the produced workpieces of the polishing robot are improved, the safety problem occurrence probability in the polishing process is reduced, and the technical effect of accurate control is realized.
In view of the above, the present application provides a flexible constant force control method and system for a grinding robot.
In a first aspect, the present application provides a flexible constant force control method for a sharpening robot, wherein the method comprises: detecting and acquiring first grinding information to be ground by the first grinding robot through the constant force compensation module; inputting the first grinding information into a constant force compensation analysis model to obtain constant force compensation information; grinding constant force compensation adjustment is carried out on the first grinding robot by adopting the constant force compensation information, and the posture of the first grinding robot is detected to obtain first posture information; constructing a posture balance compensation state space according to the first posture information; acquiring various attitude compensation behavior information in the attitude balance compensation state space; constructing attitude compensation optimization fitness, and optimizing in various attitude compensation behavior information to obtain optimal attitude compensation behavior information; and performing attitude balance compensation by adopting the optimal attitude compensation behavior information, polishing, and taking the compensated attitude information as initial attitude information for performing attitude balance compensation next time.
In another aspect, the present application provides a flexible constant force control system for a grinding robot, wherein the system comprises: the first detection unit is used for detecting and acquiring first grinding information to be ground by the first grinding robot through the constant force compensation module; the first obtaining unit is used for inputting the first grinding information into a constant force compensation analysis model to obtain constant force compensation information; the second obtaining unit is used for carrying out grinding constant force compensation adjustment on the first grinding robot by adopting the constant force compensation information, detecting the posture of the first grinding robot and obtaining first posture information; the first construction unit is used for constructing an attitude balance compensation state space according to the first attitude information; a third obtaining unit, configured to obtain multiple posture compensation behavior information in the posture balance compensation state space; the second construction unit is used for constructing attitude compensation optimization fitness, and optimizing the attitude compensation optimization fitness in various attitude compensation behavior information to obtain optimal attitude compensation behavior information; and the first execution unit is used for carrying out attitude balance compensation by adopting the optimal attitude compensation behavior information, polishing and taking the compensated attitude information as initial attitude information for carrying out attitude balance compensation next time.
In a third aspect, the present application provides a flexible constant force control system for a grinding robot comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the steps of the method of any one of the first aspect.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
the method comprises the steps of acquiring first grinding information to be ground by a first grinding robot; obtaining constant force compensation information according to the first grinding information; carrying out constant grinding force compensation adjustment on the first grinding robot, and detecting the posture of the first grinding robot to obtain first posture information; constructing a posture balance compensation state space; acquiring various attitude compensation behavior information; constructing attitude compensation optimization fitness, and optimizing in various attitude compensation behavior information; the application provides a flexible constant force control method and system for the polishing robot, so that a constant force compensation analysis model is established by collecting polishing operation parameter information of the robot, constant force compensation information is generated in a targeted manner, and a method for optimizing the posture according to the compensated posture is achieved, the control precision and the quality of produced workpieces of the polishing robot are improved, the safety problem occurrence probability in the polishing process is reduced, and the technical effect of accurate control is realized.
The foregoing description is only an overview of the technical solutions of the present application, and the present application can be implemented according to the content of the description in order to make the technical means of the present application more clearly understood, and the following detailed description of the present application is given in order to make the above and other objects, features, and advantages of the present application more clearly understandable.
Drawings
Fig. 1 is a schematic flow chart of a method for controlling a flexible constant force for a polishing robot according to an embodiment of the present disclosure;
fig. 2 is a schematic flow chart of a method for controlling a flexible constant force for a polishing robot according to an embodiment of the present disclosure to obtain constant force compensation information;
fig. 3 is a schematic flow chart of construction of attitude compensation optimization fitness of a flexible constant force control method for a polishing robot according to an embodiment of the present disclosure;
FIG. 4 is a schematic structural diagram of a flexible constant force control system for a grinding robot according to an embodiment of the present disclosure;
fig. 5 is a schematic structural diagram of an exemplary electronic device according to an embodiment of the present application.
Description of reference numerals: the system comprises a first detection unit 11, a first obtaining unit 12, a second obtaining unit 13, a first constructing unit 14, a third obtaining unit 15, a second constructing unit 16, a first execution unit 17, an electronic device 300, a memory 301, a processor 302, a communication interface 303 and a bus architecture 304.
Detailed Description
The application provides a flexible constant force control method and system for a polishing robot, and solves the technical problems that in the prior art, the precision of a position control method adopted in the polishing operation process of the robot is low, the polishing quality of a workpiece is poor, and certain hidden danger exists in the safety of the polishing robot and the workpiece. The method for acquiring the polishing operation parameter information of the robot, constructing the constant force compensation analysis model, generating the constant force compensation information in a targeted manner and optimizing the posture according to the compensated posture is achieved, the control precision and the quality of the produced workpieces of the polishing robot are improved, the safety problem occurrence probability in the polishing process is reduced, and the technical effect of accurate control is realized.
According to the technical scheme, the data acquisition, storage, use, processing and the like meet relevant regulations of national laws and regulations.
The traditional robot only has the displacement concept in the grinding and polishing operation, but for the contact operation such as polishing, grinding, assembling and the like, the requirements on the processing precision and the processing quality cannot be met only by adopting position control. In order to ensure the processing quality of a workpiece during grinding and polishing and prevent the robot or the workpiece from being damaged when the end effector is in contact with the workpiece, the robot needs to be effectively controlled by force. The technical problems that the precision of the position control method adopted in the robot polishing operation process is low, the workpiece polishing quality is poor, and the safety of the polishing robot and the workpiece has certain hidden danger exist.
In view of the above technical problems, the technical solution provided by the present application has the following general idea:
the application provides a flexible constant force control method for a grinding robot, wherein the method comprises the following steps: acquiring first grinding information to be ground by a first grinding robot; obtaining constant force compensation information according to the first grinding information; carrying out constant grinding force compensation adjustment on the first grinding robot, and detecting the posture of the first grinding robot to obtain first posture information; constructing a posture balance compensation state space; acquiring various attitude compensation behavior information; constructing attitude compensation optimization fitness, and optimizing in various attitude compensation behavior information; and carrying out attitude balance compensation by adopting the optimal attitude compensation behavior information for polishing, and taking the compensated attitude information as initial attitude information for carrying out attitude balance compensation next time. The method has the advantages that the constant force compensation analysis model is constructed by collecting the polishing operation parameter information of the robot, the constant force compensation information is generated in a targeted manner, and the attitude optimization is carried out according to the compensated attitude, so that the control precision and the quality of the produced workpieces of the polishing robot are improved, the safety problem occurrence probability in the polishing process is reduced, and the technical effect of accurate control is realized.
Having thus described the general principles of the present application, various non-limiting embodiments thereof will now be described in detail with reference to the accompanying drawings.
Example one
As shown in fig. 1, the present application provides a flexible constant force control method for a polishing robot, wherein the method is applied to a flexible constant force control system for a polishing robot, the system includes a constant force compensation module and a nine-axis attitude compensation module, and the method includes:
step S100: detecting and acquiring first grinding information to be ground by the first grinding robot through the constant force compensation module;
specifically, the conventional robot has only a displacement concept in the polishing operation, and in order to ensure the processing quality of the workpiece during polishing and prevent the robot or the workpiece from being damaged when the end effector is in contact with the workpiece, effective force control needs to be performed on the robot. The embodiment of the application therefore provides a constant force control method, which is applied to a flexible constant force control system for a grinding robot. The system comprises a constant force compensation module and a nine-axis attitude compensation module. The constant force compensation module can provide different compensation constant forces, keeps the robot to carry out constant force polishing, and the nine-axis attitude compensation module can carry out attitude acquisition in real time and keeps the polishing attitude of the robot stable.
The first grinding robot is any robot used for high-precision workpiece grinding tasks. The constant force compensation module is used for detecting the first grinding robot in real time to obtain first grinding information which is prepared by the first grinding robot for grinding, wherein the first grinding information comprises specific parameter information in grinding work such as specific workpieces to be ground, grinding routes, force requirements and grinding precision. The first grinding information can also be understood as self-adaptive parameter information set according to specific models and specific scenes when a same type of workpieces comprising a plurality of models are ground in a factory. The first grinding information to be ground of the first grinding robot is acquired through detection, and a foundation can be laid for mechanical grinding control of the robot.
Step S200: inputting the first grinding information into a constant force compensation analysis model to obtain constant force compensation information;
step S300: grinding constant force compensation adjustment is carried out on the first grinding robot by adopting the constant force compensation information, and the posture of the first grinding robot is detected to obtain first posture information;
specifically, the constant force compensation analysis model is a model trained and constructed for a specific polishing environment of the first polishing robot, and exemplarily includes: the method is obtained by training through collecting historical polishing information and corresponding historical constant force compensation information on the basis of a neural network model. And after the first grinding information is input into the constant force compensation analysis model, the constant force compensation analysis model outputs the constant force compensation information through the operation processing of internal neurons.
The constant force compensation information is used for carrying out polishing constant force compensation adjustment on the first polishing robot, namely the constant force compensation information is used for adjusting in real time in the polishing process so as to support constant force polishing, and therefore a foundation is laid for guaranteeing the uniformity of products in the polishing process.
After the constant force compensation is completed, the posture of the first polishing robot may be affected in the compensation process, and in order to ensure the balance of the robot in the polishing process, posture stable compensation is required.
And further detecting the attitude of the first grinding robot, and acquiring a multi-dimensional attitude information set of the first grinding robot, namely the first attitude information, through the nine-axis attitude compensation module. The nine-axis attitude compensation module comprises nine-axis attitude sensors (three-axis acceleration, three-axis angular velocity and three-axis gyroscope sensors), and the first attitude information is attitude information acquired through the nine-axis attitude sensor block and comprises a three-axis acceleration attitude information set, a three-axis angular velocity attitude information set and a three-axis angular motion attitude information set. And a foundation is laid for subsequent posture compensation through the acquisition of the posture information.
Step S400: constructing a posture balance compensation state space according to the first posture information;
step S500: acquiring various attitude compensation behavior information in the attitude balance compensation state space;
specifically, the attitude balance compensation space is a set including all possible attitude balance compensation behaviors for compensating the attitude under the attitude corresponding to the first attitude information, and the attitude balance compensation space is constructed according to the first attitude information, that is, a space for balancing and compensating the attitude is constructed according to the first attitude information, which is as follows: can be in a chain structure, such as: the attitude is taken as a node state, the attitude compensation is taken as an action, different attitude compensation actions can generate different states for the attitude node, namely different compensated states are generated to reach a new state, so that a sequence is formed, and various attitude compensation action information is obtained. Therefore, by implementing the compensation behavior in the attitude balance compensation state space, a variety of attitude compensation behavior information can be obtained.
Step S600: constructing attitude compensation optimization fitness, and optimizing in various attitude compensation behavior information to obtain optimal attitude compensation behavior information;
further, as shown in fig. 3, the constructing the attitude compensation optimization fitness further includes, in step S600 in the embodiment of the present application:
step S610: acquiring second attitude information after acquiring different attitude compensation behavior information and performing attitude balance compensation;
step S620: acquiring constant force control influence information on the first grinding robot after acquiring different attitude compensation behavior information and performing attitude balance compensation;
step S630: according to the severity of the polishing influence, carrying out weight distribution on the second attitude information and the constant force control influence information to obtain a weight distribution result;
step S640: and taking the second attitude information, the constant force control influence information and the weight distribution result as the optimized fitness.
Specifically, after the first attitude information is balanced and compensated based on the multiple kinds of attitude compensation behavior information, different kinds of second attitude information are obtained. And carrying out attitude balance compensation on the constant force control influence information on the first grinding robot after adopting different attitude compensation behavior information through a constant force compensation module. The constant force control influence information is influence information of compensation behaviors on constant force control polishing, and influence can be positive influence or negative influence.
Each posture in the second posture information corresponds to a constant force control influence information, and the importance of posture stability is different from the constant force control degree under different polishing work during polishing. And carrying out weight distribution on the second attitude information and the constant force control influence information according to the serious condition of the polishing influence, wherein the more serious the influence condition is, namely the larger the absolute value of positive influence and negative influence is, the larger the distributed weight is, so that the weight distribution result is obtained. And further taking the second attitude information, the constant force control influence information and the weight distribution result as the optimization fitness, optimizing various attitude compensation behavior information to obtain the optimal attitude compensation behavior, wherein the optimization can be optimized by adopting a simulated annealing optimization algorithm.
In the optimization process of the simulated annealing algorithm, the local optimal solution can probabilistically jump out and finally tend to the global optimal, so that the optimization algorithm of the serial structure which is trapped in local minimum and finally tends to the global optimal can be effectively avoided. And constructing attitude compensation optimization fitness, and optimizing in various attitude compensation behavior information through an optimization algorithm so as to obtain optimal attitude compensation behavior information and continuously search optimal attitude balance compensation for real-time adjustment to provide support.
Step S700: and performing attitude balance compensation by adopting the optimal attitude compensation behavior information, polishing, and taking the compensated attitude information as initial attitude information for performing attitude balance compensation next time.
Specifically, since the first polishing robot continuously changes the polishing attitude during the operation, after the attitude balance compensation is performed by using the optimal attitude compensation behavior information, the attitude balance compensation needs to be continuously performed after a stage of polishing task is completed, and then the compensated attitude information is used as the initial attitude information of the next attitude balance compensation, and the attitude optimization is continuously performed. Therefore, the technical effects of carrying out accurate posture control on the polishing robot and improving the polishing quality of the polishing robot through posture balance compensation and real-time optimization are achieved.
Further, as shown in fig. 2, step S200 in the embodiment of the present application further includes:
step S210: acquiring first polishing scene information, wherein the first polishing scene information comprises a plurality of kinds of polishing information, and the first polishing information is included in the plurality of kinds of polishing information;
step S220: constructing the constant force compensation analysis model according to the first grinding scene information based on a neural network model;
step S230: inputting the first grinding information into the constant force compensation analysis model to obtain a first output result;
step S240: acquiring feedforward constant force compensation information and feedback constant force compensation information according to the first output result;
step S250: and taking the feedforward constant force compensation information and the feedback constant force compensation information as the constant force compensation information.
Further, the step S220 of the embodiment of the present application further includes, based on the neural network model, constructing the constant force compensation analysis model according to the first grinding scenario information:
step S221: constructing an input layer, a hidden layer and an output layer of the constant force compensation analysis model based on a neural network model, wherein the hidden layer comprises a feedforward constant force compensation analysis network and a feedback constant force compensation analysis network;
step S222: acquiring a plurality of groups of feedforward constant force compensation data, wherein each group of feedforward constant force compensation data comprises polishing information, initial polishing constant force information, feedforward constant force compensation information and adjusted polishing constant force information;
step S223: dividing and identifying a plurality of groups of feedforward constant force compensation data, and performing supervision training, verification and test on the feedforward constant force compensation analysis network until the accuracy of the feedforward constant force compensation analysis network reaches a preset requirement;
step S224: acquiring a plurality of groups of feedback constant force compensation data, wherein each group of feedback constant force compensation data comprises polishing information, adjusted polishing constant force information and feedback constant force compensation information;
step S225: dividing and identifying a plurality of groups of feedback constant force compensation data, and performing supervision training, verification and test on the feedback constant force compensation analysis network until the accuracy of the feedback constant force compensation analysis network reaches a preset requirement;
step S226: the feedforward constant force compensation analysis network and the feedback constant force compensation analysis network are connected in a full mode, and combined training is carried out;
step S227: and obtaining the constant force compensation analysis model.
Specifically, in order to perform accurate constant force compensation on the first polishing robot, training of a constant force compensation analysis model is performed on first polishing scene information. The method comprises the steps of collecting a plurality of kinds of polishing information in a first polishing information scene through means such as big data or image collection, wherein the plurality of kinds of polishing information comprise the first polishing information, namely the first polishing information is one of the plurality of kinds of polishing information.
Constructing the constant force compensation analysis model based on the first grinding scene, namely training the neural network model through the first grinding scene information, wherein the preferable method comprises the following steps: firstly, a model network layer is constructed, wherein the model network layer comprises an input layer, a hidden layer and an output layer, and the hidden layer comprises a feedforward constant force compensation analysis network and a feedback constant force compensation analysis network. The feedforward constant force compensation analysis network is used for feedforward intervention, namely feedforward constant force compensation, and the feedback constant force compensation analysis network is used for feedback intervention, namely feedback constant force compensation. And further respectively training a feedforward constant force compensation analysis network layer and a feedback constant force compensation analysis network layer.
Specifically, the grinding information, the initial grinding constant force information, the feedforward constant force compensation information and the adjustment grinding constant force information are collected and used as a plurality of groups of feedforward constant force compensation data, wherein the grinding information and the initial grinding constant force information are input parameters, the feedforward constant force compensation information and the adjustment grinding constant force information are output parameters, and the adjustment grinding constant force information is constant force information compensated based on the feedforward constant force compensation information. And dividing the plurality of groups of feedforward constant force compensation data into a training set, a testing set and a verification set, further identifying the plurality of groups of feedforward constant force compensation information, carrying out supervision training, verification and testing on the feedforward constant force compensation analysis network until the accuracy of the feedforward constant force compensation analysis network reaches a preset requirement, and stopping training to obtain the feedforward constant force compensation analysis network.
And the grinding information is collected, the grinding constant force information is adjusted, and the feedback constant force compensation information is fed back to form the multiple groups of feedback constant force compensation data. The polishing information and the adjusted polishing constant force information are used as network input parameters, and the feedback constant force compensation information is used as a network output parameter. And dividing a plurality of groups of the feedback constant force compensation data into a training set, a testing set and a verification set, further identifying the plurality of groups of the feedback constant force compensation data, carrying out supervision training, verification and testing on the feedback constant force compensation analysis network until the accuracy of the feedback constant force compensation analysis network reaches a preset requirement, and stopping training to obtain the feedback constant force compensation analysis network.
And fully connecting the feedforward constant force compensation analysis network with the feedback constant force compensation analysis network, and adjusting the polishing constant force information by using the output information of the feedforward constant force compensation analysis network as one of the input information of the feedback constant force compensation analysis network. After the two-layer network structure is fully connected, the hidden layer formed by the two network structures is trained in a joint training mode, and an unrestricted example is shown: the feedforward constant force compensation analysis network can be trained through a training data set, the feedback constant force compensation analysis network is trained by combining the adjusted grinding constant force information in the output information of the feedforward constant force compensation analysis network with other collected training data sets (grinding information and feedback constant force compensation information), and the training is stopped until the output result of the fully-connected network layer reaches a certain accuracy rate or convergence. The loss in the separate training of the feedforward constant force compensation analysis network and the feedback constant force compensation analysis network can be compensated by the combined training.
And after the constant force compensation analysis model is obtained, inputting the first grinding information into the constant force compensation analysis model to obtain a first output result, wherein the first output result comprises the feedforward constant force compensation information and the feedback constant force compensation information, and the obtained feedforward constant force compensation information and the feedback constant force compensation information are used as the constant force compensation information for carrying out grinding constant force compensation adjustment on the first grinding robot. The feed-forward constant force compensation information is a pre-adjusted larger-amplitude force predicted according to the first grinding information and is used for supporting constant force grinding. The feedback constant force compensation information is the force which is adjusted in a small-amplitude adaptability mode after the feedforward constant force compensation is simulated. Therefore, the constant force compensation analysis model is constructed in a targeted manner through the polishing environment information, the targeted generation of the constant force compensation information can be achieved, and the technical effect of improving the constant force compensation accuracy is achieved.
Further, the detecting the pose of the first grinding robot further includes, in step S300 of the embodiment of the present application:
step S310: acquiring and acquiring a multi-dimensional attitude information set of the first grinding robot through the nine-axis attitude compensation module, wherein the multi-dimensional attitude information set comprises a three-axis acceleration attitude information set, a three-axis angular velocity attitude information set and a three-axis angular motion attitude information set;
step S320: and performing fusion dimensionality reduction on the multi-dimensional attitude information set to obtain the first attitude information.
Further, the step S320 of performing fusion and dimension reduction on the multi-dimensional posture information set further includes:
step S321: performing decentralized processing on the multi-dimensional attitude information set to obtain a feature data set;
step S322: calculating to obtain a covariance matrix of the feature data set;
step S323: calculating the covariance matrix to obtain an eigenvalue and an eigenvector of the covariance matrix;
step S324: projecting the multi-dimensional attitude information set to the characteristic vector to obtain a dimension reduction data set;
step S325: and taking the dimension reduction data set as the first posture information.
Specifically, a multi-dimensional attitude information set of the first grinding robot is acquired through a sensor of the nine-axis attitude compensation module, and the multi-dimensional attitude information set includes a three-axis acceleration attitude information set, a three-axis angular velocity attitude information set and a three-axis angular motion attitude information set.
Because the acquired information set is multi-dimensional data and has higher spatial complexity for description of the attitude information, in order to reduce the information dimension, a fusion dimension reduction method is adopted to obtain the first attitude information. The fusion dimensionality reduction method can perform multiple dimensionality reduction on the multi-dimensional attitude information, for example, firstly perform dimensionality reduction on a triaxial acceleration attitude information set, a triaxial angular velocity attitude information set and a triaxial angular motion attitude information set, then obtain three dimensionality reduction data sets, and then perform secondary dimensionality reduction, wherein the dimensionality reduction method can use a principal component analysis method.
Specifically, an average value is obtained in the multi-dimensional attitude information set, and an average value is removed from each sample data in the multi-dimensional attitude information set, that is, the de-centering processing is performed, so as to obtain a feature data set. The characteristic data set is a new data set with the mean value removed, a data matrix is formed, and the characteristic data set is operated through a covariance formula to obtain a first covariance matrix. And obtaining eigenvalues and eigenvectors according to the first covariance matrix, wherein the eigenvalues and the eigenvectors are in one-to-one correspondence, selecting the first K eigenvalues and eigenvectors, projecting the multi-dimensional attitude information set onto the eigenvectors to obtain the dimensionality reduced data set, and using the dimensionality reduced data set as the first attitude information after dimensionality reduction.
The multidimensional attitude information set is subjected to dimensionality reduction processing through a principal component analysis method, so that data dimensionality reduction is achieved on the premise of ensuring enough information quantity, minimum data loss can be ensured, and the speed of data operation is increased.
Further, the optimizing is performed in the multiple kinds of posture compensation behavior information, and step S600 in the embodiment of the present application further includes:
step S641: randomly selecting attitude compensation behavior information from the various attitude compensation behavior information as first attitude compensation behavior information and as an optimal solution;
step S642: calculating the fitness of the first attitude compensation behavior information according to the optimized fitness to obtain a first fitness;
step S643: randomly selecting attitude compensation behavior information from the plurality of types of attitude compensation behavior information as second attitude compensation behavior information;
step S644: calculating the fitness of the second attitude compensation behavior information according to the optimized fitness to obtain a second fitness;
step S645: if the second fitness is greater than the first fitness, replacing the first posture compensation behavior information with the second posture compensation behavior information to serve as the optimal solution;
step S646: and if the second fitness is smaller than the first fitness, replacing the first posture compensation behavior information with the second posture compensation behavior information according to a probability as the optimal solution, wherein the probability is calculated by the following formula:
wherein r is 2 Is a second fitness, r 1 K is an optimized speed factor;
step S647: if the optimal solution is not changed in the iterative optimization of the threshold times, outputting the optimal solution, or if the iterative optimization reaches a preset time, outputting the optimal solution to obtain the optimal attitude compensation behavior information.
Specifically, the specific optimization process performed within the plurality of types of attitude compensation behavior information is as follows, first randomly selecting and setting any one of the attitude compensation behavior information as the first attitude compensation behavior information, and using the first attitude compensation behavior information as the spatial optimal solution. And calculating the fitness of the first attitude compensation behavior according to the optimized fitness, wherein the optimized fitness can be understood as the fitness of the compensation behavior after attitude balance compensation, and the better the fitness is, the more accurate the compensation behavior is. In the specific calculation process, the corresponding compensated second attitude information and the compensated constant force control influence information on the constant force control can be obtained according to the first attitude compensation behavior, and the weighted calculation is further carried out according to the weight distribution result to obtain the corresponding first fitness. Randomly selecting one attitude compensation behavior information from the plurality of attitude compensation behavior information as second attitude compensation behavior information, and analyzing according to the fitness to obtain the fitness of the second attitude compensation behavior information, namely the second fitness.
And comparing the second fitness with the first fitness, and replacing the first posture compensation behavior information with the second posture compensation behavior information as an optimal solution if the second fitness is greater than the first fitness. In addition, if the second fitness is smaller than the first fitness, the acceptance probability needs to be determined according to a formula
The calculation of the acceptance probability is performed. Wherein r is 2 Is a second fitness, r 1 For the first fitness, k is an optimized speed factor, so that the acceptance probability is related to the difference value of the first fitness and the second fitness. k is a constant which is gradually reduced along with the optimization iteration number, at the initial stage of optimization, k is larger, the probability of first posture compensation behavior information is not the globally optimal posture compensation behavior information, and may be locally optimal, in order to avoid the optimization process from being stopped at the locally optimal position, k is larger, so that P is larger, the probability of accepting inferior second posture compensation behavior information is larger as an optimal solution, at the later stage of optimization, the probability of current optimal posture compensation behavior information may be the globally optimal posture compensation behavior information, in order to improve the accuracy of optimization, k is smaller, so that P is smaller, the probability of accepting inferior compensation behavior information is smaller as the globally optimal compensation behavior information, and the accuracy of optimization is improved. OptionallyThe decreasing manner of k may be any decreasing manner in the prior art, such as exponential decreasing or logarithmic decreasing, and the value of k and the decreasing manner may be determined according to the amount of the compensation behavior information.
Further, a threshold number of times is set, the threshold number of times is limited by the number of iterations, and if the iterations are repeated for multiple times until the threshold number of times is met, the optimal solution is not changed and is output as the optimal solution. Or when the iterative optimization times reach preset times, outputting the optimal solution to obtain the optimal attitude compensation behavior information. The method has the technical effects of determining the optimal solution in various attitude compensation behavior information, improving the reliability of attitude compensation and improving the optimization efficiency by setting the optimization fitness, carrying out multiple iterations in an optimization space, carrying out fitness comparison and analyzing the acceptance probability.
In summary, the flexible constant force control method and system for the polishing robot provided by the embodiment of the application have the following technical effects:
1. the method comprises the steps of acquiring first grinding information to be ground by a first grinding robot; obtaining constant force compensation information according to the first grinding information; carrying out constant grinding force compensation adjustment on the first grinding robot, and detecting the posture of the first grinding robot to obtain first posture information; constructing an attitude balance compensation state space; acquiring various attitude compensation behavior information; constructing attitude compensation optimization fitness, and optimizing in various attitude compensation behavior information; the embodiment of the application provides a flexible constant force control method and system for a polishing robot, so that a constant force compensation analysis model is established by collecting polishing operation parameter information of the robot, constant force compensation information is generated in a targeted manner, and a method for optimizing the posture according to the compensated posture is achieved, the control precision and the quality of a produced workpiece of the polishing robot are improved, the safety problem occurrence probability in the polishing process is reduced, and the technical effect of accurate control is realized.
2. By setting the optimization fitness, carrying out multiple iterations in the optimization space, comparing the fitness and analyzing the acceptance probability, the technical effects of determining the optimal solution in various attitude compensation behavior information, improving the reliability of attitude compensation and improving the optimization efficiency are achieved.
3. The multidimensional attitude information set is subjected to dimensionality reduction processing through a principal component analysis method, so that the technical effects of performing data dimensionality reduction on the premise of ensuring enough information quantity, ensuring minimum data loss and improving the speed of data operation are achieved.
Example two
Based on the same inventive concept as the flexible constant force control method for the grinding robot in the previous embodiment, as shown in fig. 4, the present embodiment provides a flexible constant force control system for the grinding robot, wherein the system comprises:
the first detection unit 11 is used for detecting and acquiring first grinding information to be ground by the first grinding robot through the constant force compensation module;
the first obtaining unit 12 is configured to input the first grinding information into a constant force compensation analysis model, so as to obtain constant force compensation information;
a second obtaining unit 13, where the second obtaining unit 13 is configured to perform constant force compensation adjustment on the first polishing robot by using the constant force compensation information, and detect a posture of the first polishing robot to obtain first posture information;
a first constructing unit 14, wherein the first constructing unit 14 is configured to construct an attitude balance compensation state space according to the first attitude information;
a third obtaining unit 15, where the third obtaining unit 15 is configured to obtain multiple posture compensation behavior information in the posture balance compensation state space;
the second construction unit 16, the second construction unit 16 is configured to construct an attitude compensation optimization fitness, and perform optimization within the plurality of types of attitude compensation behavior information to obtain optimal attitude compensation behavior information;
and the first execution unit 17 is configured to perform attitude balance compensation by using the optimal attitude compensation behavior information, perform polishing, and use the compensated attitude information as initial attitude information for performing attitude balance compensation next time.
Further, the system comprises:
the first acquisition unit is used for acquiring and acquiring first polishing scene information, wherein the first polishing scene information comprises a plurality of kinds of polishing information, and the first polishing information is contained in the plurality of kinds of polishing information;
the third construction unit is used for constructing the constant force compensation analysis model according to the first grinding scene information based on a neural network model;
a fourth obtaining unit, configured to input the first grinding information into the constant force compensation analysis model to obtain a first output result;
a fifth obtaining unit, configured to obtain feedforward constant force compensation information and feedback constant force compensation information according to the first output result;
a second execution unit, configured to use the feedforward constant force compensation information and the feedback constant force compensation information as the constant force compensation information.
Further, the system comprises:
a fourth construction unit, configured to construct an input layer, a hidden layer, and an output layer of the constant force compensation analysis model based on a neural network model, where the hidden layer includes a feedforward constant force compensation analysis network and a feedback constant force compensation analysis network;
the second acquisition unit is used for acquiring and acquiring a plurality of groups of feedforward constant force compensation data, wherein each group of feedforward constant force compensation data comprises polishing information, initial polishing constant force information, feedforward constant force compensation information and adjusted polishing constant force information;
the third execution unit is used for dividing and identifying a plurality of groups of feedforward constant force compensation data, and performing supervision training, verification and test on the feedforward constant force compensation analysis network until the accuracy of the feedforward constant force compensation analysis network reaches a preset requirement;
the third acquisition unit is used for acquiring and acquiring a plurality of groups of feedback constant force compensation data, wherein each group of feedback constant force compensation data comprises polishing information, adjusted polishing constant force information and feedback constant force compensation information;
the fourth execution unit is used for dividing and identifying a plurality of groups of feedback constant force compensation data, and performing supervision training, verification and test on the feedback constant force compensation analysis network until the accuracy of the feedback constant force compensation analysis network reaches a preset requirement;
the fifth execution unit is used for being fully connected with the feedforward constant force compensation analysis network and the feedback constant force compensation analysis network and performing combined training;
a sixth obtaining unit, configured to obtain the constant force compensation analysis model.
Further, the system comprises:
a fourth acquisition unit, configured to acquire and acquire a multi-dimensional attitude information set of the first grinding robot through the nine-axis attitude compensation module, where the multi-dimensional attitude information set includes a three-axis acceleration attitude information set, a three-axis angular velocity attitude information set, and a three-axis angular motion attitude information set;
a seventh obtaining unit, configured to perform fusion dimensionality reduction on the multi-dimensional pose information set to obtain the first pose information.
Further, the system comprises:
an eighth obtaining unit, configured to perform decentralized processing on the multi-dimensional attitude information set to obtain a feature data set;
a sixth execution unit, configured to calculate a covariance matrix of the feature data set;
a ninth obtaining unit, configured to perform operation on the covariance matrix to obtain an eigenvalue and an eigenvector of the covariance matrix;
a tenth obtaining unit, configured to project the multi-dimensional pose information set onto the feature vector to obtain a dimension-reduced dataset;
a seventh execution unit to treat the dimension reduction dataset as the first pose information.
Further, the system comprises:
the fifth acquisition unit is used for acquiring second attitude information obtained by performing attitude balance compensation on different attitude compensation behavior information;
the sixth acquisition unit is used for acquiring constant force control influence information on the first grinding robot after acquiring different attitude compensation behavior information to perform attitude balance compensation;
an eleventh obtaining unit, configured to perform weight distribution on the second posture information and the constant force control influence information according to a severity of a polishing influence, and obtain a weight distribution result;
an eighth execution unit, configured to use the second posture information, the constant force control influence information, and the weight distribution result as the optimal fitness.
Further, the system comprises:
a ninth execution unit, configured to randomly select one attitude compensation behavior information from among the plurality of types of attitude compensation behavior information, as the first attitude compensation behavior information, and as an optimal solution;
a twelfth obtaining unit, configured to calculate a fitness of the first posture compensation behavior information according to the optimized fitness to obtain a first fitness;
a tenth execution unit, configured to randomly select one attitude compensation behavior information from among the plurality of types of attitude compensation behavior information, as second attitude compensation behavior information;
a thirteenth obtaining unit, configured to calculate a fitness of the second posture compensation behavior information according to the optimized fitness to obtain a second fitness;
an eleventh execution unit, configured to replace the first posture compensation behavior information with the second posture compensation behavior information as the optimal solution if the second fitness is greater than the first fitness;
a first calculating unit, configured to, if the second fitness is smaller than the first fitness, replace the first posture compensation behavior information with the second posture compensation behavior information according to a probability as the optimal solution, where the probability is calculated by the following equation:
wherein r is 2 Is a second fitness, r 1 K is an optimized speed factor for the first fitness;
a fourteenth obtaining unit, configured to, if the optimal solution does not change in the threshold number of iterative optimizations, output the optimal solution, or, if the iterative optimization reaches a preset number of times, output the optimal solution, and obtain the optimal attitude compensation behavior information.
Exemplary electronic device
The electronic device of the embodiment of the present application is described below with reference to fig. 5.
Based on the same inventive concept as the flexible constant force control method for the polishing robot in the foregoing embodiments, the present application also provides a flexible constant force control system for the polishing robot, including: a processor coupled to a memory, the memory for storing a program that, when executed by the processor, causes the system to perform the method of embodiment one.
The electronic device 300 includes: processor 302, communication interface 303, memory 301. Optionally, the electronic device 300 may also include a bus architecture 304. Wherein, the communication interface 303, the processor 302 and the memory 301 may be connected to each other through a bus architecture 304; the bus architecture 304 may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus architecture 304 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 5, but this is not intended to represent only one bus or type of bus.
The communication interface 303 is a system using any transceiver or the like, and is used for communicating with other devices or communication networks, such as ethernet, radio Access Network (RAN), wireless Local Area Network (WLAN), wired access network, and the like.
The memory 301 may be, but is not limited to, a ROM or other type of static storage device that can store static information and instructions, a RAM or other type of dynamic storage device that can store information and instructions, an electrically erasable Programmable read-only memory (EEPROM), a compact disk read-only memory (CD-ROM) or other optical disk storage, optical disk storage (including compact disk, laser disk, optical disk, digital versatile disk, blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. The memory may be self-contained and coupled to the processor through a bus architecture 304. The memory may also be integrated with the processor.
The memory 301 is used for storing computer-executable instructions for executing the present application, and is controlled by the processor 302 to execute. The processor 302 is used for executing computer-executable instructions stored in the memory 301, so as to implement a flexible constant force control method for a polishing robot provided by the above embodiments of the present application.
Optionally, the computer-executable instructions in the embodiments of the present application may also be referred to as application program codes, which are not specifically limited in the embodiments of the present application.
The embodiment of the application provides a flexible constant force control method for a polishing robot, wherein the method comprises the following steps: acquiring first grinding information to be ground by a first grinding robot; obtaining constant force compensation information according to the first grinding information; carrying out constant grinding force compensation adjustment on the first grinding robot, and detecting the posture of the first grinding robot to obtain first posture information; constructing a posture balance compensation state space; acquiring various attitude compensation behavior information; constructing attitude compensation optimization fitness, and optimizing in various attitude compensation behavior information; and carrying out attitude balance compensation by adopting the optimal attitude compensation behavior information for polishing, and taking the compensated attitude information as initial attitude information for carrying out attitude balance compensation next time.
Those of ordinary skill in the art will understand that: the various numbers of the first, second, etc. mentioned in this application are only used for the convenience of description and are not used to limit the scope of the embodiments of this application, nor to indicate the order of precedence. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one" means one or more. At least two means two or more. "at least one," "any," or similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one (one ) of a, b, or c, may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or multiple.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the application to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable system. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device including one or more available media integrated servers, data centers, and the like. The usable medium may be a magnetic medium (e.g., a floppy Disk, a hard Disk, a magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a Solid State Disk (SSD)), among others.
The various illustrative logical units and circuits described in this application may be implemented or operated upon by general purpose processors, digital signal processors, application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other programmable logic systems, discrete gate or transistor logic, discrete hardware components, or any combination thereof. A general-purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing systems, e.g., a digital signal processor and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a digital signal processor core, or any other similar configuration.
The steps of a method or algorithm described in the embodiments herein may be embodied directly in hardware, in a software element executed by a processor, or in a combination of the two. The software cells may be stored in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. For example, a storage medium may be coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC, which may be disposed in a terminal. In the alternative, the processor and the storage medium may reside in different components within the terminal. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Although the present application has been described in conjunction with specific features and embodiments thereof, it will be evident that various modifications and combinations can be made thereto without departing from the spirit and scope of the application. Accordingly, the specification and figures are merely exemplary of the present application as defined herein, and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of the present application. It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the scope of the application. Thus, it is intended that the present application include such modifications and variations as come within the scope of the application and its equivalents.
Claims (8)
1. A flexible constant force control method for a polishing robot, the method being applied to a flexible constant force control system for a polishing robot, the system including a constant force compensation module and a nine-axis attitude compensation module, the method comprising:
detecting and acquiring first grinding information to be ground by the first grinding robot through the constant force compensation module;
inputting the first grinding information into a constant force compensation analysis model to obtain constant force compensation information;
grinding constant force compensation adjustment is carried out on the first grinding robot by adopting the constant force compensation information, and the posture of the first grinding robot is detected to obtain first posture information;
constructing a posture balance compensation state space according to the first posture information;
acquiring various attitude compensation behavior information in the attitude balance compensation state space;
constructing attitude compensation optimization fitness, optimizing in various attitude compensation behavior information to obtain optimal attitude compensation behavior information, wherein the constructing attitude compensation optimization fitness comprises the following steps: acquiring second attitude information after acquiring different attitude compensation behavior information and performing attitude balance compensation; acquiring constant force control influence information on the first grinding robot after acquiring different attitude compensation behavior information and performing attitude balance compensation; according to the severity of the polishing influence, carrying out weight distribution on the second attitude information and the constant force control influence information to obtain a weight distribution result; taking the second attitude information, the constant force control influence information and the weight distribution result as the optimized fitness;
and performing attitude balance compensation by adopting the optimal attitude compensation behavior information, polishing, and taking the compensated attitude information as initial attitude information for performing attitude balance compensation next time.
2. The method of claim 1, wherein said inputting the first sanding information into a constant force compensation analysis model comprises:
acquiring first polishing scene information, wherein the first polishing scene information comprises a plurality of kinds of polishing information, and the first polishing information is included in the plurality of kinds of polishing information;
constructing the constant force compensation analysis model according to the first grinding scene information based on a neural network model;
inputting the first grinding information into the constant force compensation analysis model to obtain a first output result;
acquiring feedforward constant force compensation information and feedback constant force compensation information according to the first output result;
and taking the feedforward constant force compensation information and the feedback constant force compensation information as the constant force compensation information.
3. The method of claim 2, wherein the constructing the constant force compensation analysis model from the first grinding scenario information based on the neural network model comprises:
constructing an input layer, a hidden layer and an output layer of the constant force compensation analysis model based on a neural network model, wherein the hidden layer comprises a feedforward constant force compensation analysis network and a feedback constant force compensation analysis network;
acquiring a plurality of groups of feedforward constant force compensation data, wherein each group of feedforward constant force compensation data comprises polishing information, initial polishing constant force information, feedforward constant force compensation information and adjusted polishing constant force information;
dividing and identifying a plurality of groups of feedforward constant force compensation data, and performing supervision training, verification and test on the feedforward constant force compensation analysis network until the accuracy of the feedforward constant force compensation analysis network reaches a preset requirement;
acquiring a plurality of groups of feedback constant force compensation data, wherein each group of feedback constant force compensation data comprises polishing information, adjusted polishing constant force information and feedback constant force compensation information;
dividing and identifying a plurality of groups of feedback constant force compensation data, and performing supervision training, verification and test on the feedback constant force compensation analysis network until the accuracy of the feedback constant force compensation analysis network reaches a preset requirement;
the feedforward constant force compensation analysis network and the feedback constant force compensation analysis network are connected in a full mode, and combined training is carried out;
and obtaining the constant force compensation analysis model.
4. The method of claim 1 or 2, wherein said detecting a pose of said first grinding robot comprises:
acquiring and acquiring a multi-dimensional attitude information set of the first grinding robot through the nine-axis attitude compensation module, wherein the multi-dimensional attitude information set comprises a three-axis acceleration attitude information set, a three-axis angular velocity attitude information set and a three-axis angular motion attitude information set;
and performing fusion dimensionality reduction on the multi-dimensional attitude information set to obtain the first attitude information.
5. The method according to claim 4, wherein the performing fusion dimensionality reduction on the multi-dimensional pose information set comprises:
performing decentralized processing on the multi-dimensional attitude information set to obtain a feature data set;
calculating to obtain a covariance matrix of the characteristic data set;
calculating the covariance matrix to obtain an eigenvalue and an eigenvector of the covariance matrix;
projecting the multi-dimensional attitude information set to the characteristic vector to obtain a dimension reduction data set;
and taking the dimension reduction data set as the first posture information.
6. The method of claim 1, wherein the optimizing within the plurality of attitude compensation behavior information comprises:
randomly selecting attitude compensation behavior information from the various attitude compensation behavior information as first attitude compensation behavior information and as an optimal solution;
calculating the fitness of the first attitude compensation behavior information according to the optimized fitness to obtain a first fitness;
randomly selecting attitude compensation behavior information from the plurality of types of attitude compensation behavior information as second attitude compensation behavior information;
calculating the fitness of the second attitude compensation behavior information according to the optimized fitness to obtain a second fitness;
if the second fitness is greater than the first fitness, replacing the first posture compensation behavior information with the second posture compensation behavior information to serve as the optimal solution;
and if the second fitness is smaller than the first fitness, replacing the first posture compensation behavior information with the second posture compensation behavior information according to a probability as the optimal solution, wherein the probability is calculated by the following formula:
wherein r is 2 Is a second fitness, r 1 K is an optimized speed factor for the first fitness;
if the optimal solution is not changed in the iterative optimization of the threshold times, outputting the optimal solution, or if the iterative optimization reaches a preset time, outputting the optimal solution to obtain the optimal attitude compensation behavior information.
7. A flexible constant force control system for a grinding robot, the system comprising:
the first detection unit is used for detecting and acquiring first grinding information to be ground by the first grinding robot through the constant force compensation module;
the first obtaining unit is used for inputting the first grinding information into a constant force compensation analysis model to obtain constant force compensation information;
the second obtaining unit is used for carrying out grinding constant force compensation adjustment on the first grinding robot by adopting the constant force compensation information, detecting the posture of the first grinding robot and obtaining first posture information;
the first construction unit is used for constructing an attitude balance compensation state space according to the first attitude information;
a third obtaining unit, configured to obtain multiple posture compensation behavior information in the posture balance compensation state space;
the second construction unit is used for constructing attitude compensation optimization fitness, and optimizing the attitude compensation optimization fitness in various attitude compensation behavior information to obtain optimal attitude compensation behavior information;
the fifth acquisition unit is used for acquiring second attitude information obtained by performing attitude balance compensation on different attitude compensation behavior information;
the sixth acquisition unit is used for acquiring constant force control influence information on the first grinding robot after acquiring different attitude compensation behavior information to perform attitude balance compensation;
an eleventh obtaining unit, configured to perform weight distribution on the second posture information and the constant force control influence information according to a severity of a polishing influence, and obtain a weight distribution result;
an eighth execution unit, configured to use the second posture information, the constant force control influence information, and the weight distribution result as the optimal fitness;
and the first execution unit is used for carrying out attitude balance compensation by adopting the optimal attitude compensation behavior information, polishing and taking the compensated attitude information as initial attitude information for carrying out attitude balance compensation next time.
8. A flexible constant force control system for a grinding robot, comprising: a processor coupled to a memory for storing a program that, when executed by the processor, causes a system to perform the method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210422569.1A CN114800500B (en) | 2022-04-21 | 2022-04-21 | Flexible constant force control method and system for polishing robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210422569.1A CN114800500B (en) | 2022-04-21 | 2022-04-21 | Flexible constant force control method and system for polishing robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114800500A CN114800500A (en) | 2022-07-29 |
| CN114800500B true CN114800500B (en) | 2023-03-24 |
Family
ID=82505420
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210422569.1A Active CN114800500B (en) | 2022-04-21 | 2022-04-21 | Flexible constant force control method and system for polishing robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114800500B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115179193B (en) * | 2022-09-13 | 2022-12-06 | 季华实验室 | Self-adaptive constant-force output method, device, medium and electronic equipment |
| CN115331224B (en) * | 2022-10-17 | 2023-01-31 | 日照福瑞德科技有限公司 | Glass lens polishing quality analysis optimization method and system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107838920A (en) * | 2017-12-20 | 2018-03-27 | 芜湖哈特机器人产业技术研究院有限公司 | A kind of robot polishing Force control system and method |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE69130397T2 (en) * | 1990-02-27 | 1999-04-15 | Kabushiki Kaisha Toshiba, Kawasaki, Kanagawa | Device for controlling the machining force of a tool |
| CN105643399B (en) * | 2015-12-29 | 2018-06-26 | 沈阳理工大学 | The complex-curved automatic grinding-polishing system of robot and processing method based on Shared control |
| CN108972545B (en) * | 2018-06-22 | 2021-08-10 | 华南理工大学 | Robot constant-force curved surface tracking method based on fuzzy iterative algorithm |
| CN110814930B (en) * | 2019-10-30 | 2021-12-07 | 北京空间机电研究所 | Device and method for processing SiC material optical element |
| CN111273606B (en) * | 2020-03-09 | 2022-06-10 | 西南交通大学 | Tool posture optimization method for geometric error compensation of five-axis machine tool |
| CN113211460B (en) * | 2021-05-01 | 2022-03-15 | 吉林大学 | Device for improving machining rigidity of two single-arm industrial robots in non-contact mode |
-
2022
- 2022-04-21 CN CN202210422569.1A patent/CN114800500B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107838920A (en) * | 2017-12-20 | 2018-03-27 | 芜湖哈特机器人产业技术研究院有限公司 | A kind of robot polishing Force control system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114800500A (en) | 2022-07-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114800500B (en) | Flexible constant force control method and system for polishing robot | |
| US10692018B2 (en) | Machine learning device and machine learning method for learning optimal object grasp route | |
| Ren et al. | MCTAN: A novel multichannel temporal attention-based network for industrial health indicator prediction | |
| Wu et al. | Parametric identification and structure searching for underwater vehicle model using symbolic regression | |
| US20220019866A1 (en) | Controlling robots using entropy constraints | |
| Guo et al. | Study of static thermal deformation modeling based on a hybrid CNN-LSTM model with spatiotemporal correlation | |
| CN113591240B (en) | Modeling method for thermal error model of tooth grinding machine based on bidirectional LSTM network | |
| WO2020149971A2 (en) | Robust and data-efficient blackbox optimization | |
| CN112959326B (en) | Method and device for solving positive kinematics of robot, readable storage medium and robot | |
| CN108846200B (en) | Quasi-static bridge influence line identification method based on iteration method | |
| CN109740052A (en) | The construction method and device of network behavior prediction model, network behavior prediction technique | |
| CN105509748B (en) | The air navigation aid and device of robot | |
| JP2017120587A (en) | Machine learning device learning cache lock, industrial machinery system, manufacturing system, machine learning method, and machine learning program | |
| CN113156473B (en) | Self-adaptive judging method for satellite signal environment of information fusion positioning system | |
| Pretorius et al. | Evolutionary robotics applied to hexapod locomotion: A comparative study of simulation techniques | |
| CN108959787B (en) | Thermal deformation prediction method and system of macro-macro dual-drive system considering actual working conditions | |
| Messay et al. | GPGPU acceleration of a novel calibration method for industrial robots | |
| Feng et al. | Mobile robot obstacle avoidance based on deep reinforcement learning | |
| CN114580624A (en) | Method, apparatus, and computer-readable storage medium for training neural network | |
| Yetki et al. | Optimal design of deformation monitoring networks using PSO algorithm | |
| CN117406759B (en) | Pipeline robot crawling calibration method and system | |
| CN117348577B (en) | Production process simulation detection method, device, equipment and medium | |
| CN113485107B (en) | Reinforced learning robot control method and system based on consistency constraint modeling | |
| CN116051633B (en) | 3D point cloud target detection method and device based on weighted relation perception | |
| KR102441086B1 (en) | Machine learning method for compensating the imbalance of training data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |